Over 90% of all HIV infections occur in developing countries. In this project the specific objectives are to define the unique epidemiologic, clinical, virologic, and immunologic features of HIV infection in developing countries; to determine the viral kinetics associated with perinatal and heterosexual transmission and to characterize the molecular strains of HIV throughout the world for infectiousness and the immunologic response to cross-clade vaccines. In collaborative studies we have established cohorts of high-risk individuals in India, Uganda, Republic of South Africa, Brazil, and the U.S., and have characterized the prevalence, incidence, and risk behaviors for HIV infection in each respective cohort. We completed a four-year trial of mass antibiotic treatment of STDs in Uganda, which demonstrated a marked reduction in STDs but without any demonstrable effect on HIV incidence due to the low population attributable fraction associated with STDs and HIV. From this study, 415 couples who are serodiscordant for HIV infection were identified, and viral load in the infected partner was found to be the major predictor of heterosexual transmission. Although the rate of male-to-female transmission was similar to female-to-male transmission, male circumcision was protective and associated with decreased acquisition of HIV infection. In several perinatal cohort studies, we demonstrated that blood and breast milk viral load was significantly associated with increased perinatal transmission. Our laboratory described the viral kinetics of HIV in a cohort of infected children from birth to five years of age, demonstrating the predictive value of a high viral load with a rapid progression, and the clinical utility of early diagnosis based on viral RNA. We also observed a gender difference in HIV viral levels and viral dynamics. Women have half the viral level as men from time of seroconversion through four years of infection, but viral levels then increase rapidly. Further study of these women demonstrated different viral dynamics with increasing viral load trajectories in women compared to men suggesting the possibility of hormonal influence on viral replication. The significance of these studies is that they provide important epidemiologic, clinical, virologic and immunologic knowledge of HIV infection in developing countries as well as in the U.S., which can be utilized for monitoring future trends of the epidemic and for developing behavioral and biological interventions to prevent further transmission.